Top AI Repos — open-source AI, indexed and scored
Top AI Repos tracks AI repositories on GitHub and answers two different questions about each one: is it moving right now, and would you bet a product on it.
Top AI Repos tracks AI repositories on GitHub and answers two different questions about each one: is it moving right now, and would you bet a product on it.
A multi-llm Emacs shell (ChatGPT, Claude, DeepSeek, Gemini, Kagi, Ollama, Perplexity) + editing integrations
| Date | Stars |
|---|---|
| 2026-07-31 | 1214 |
| 2026-08-01 | 1214 |
| 2026-08-02 | 1214 |
| 2026-08-06 | 1214 |
Today
— stars today
This week
— stars this week
This month
— stars this month
Momentum
0.0
growth rate 0.00%/day
👉 [[https://github.com/sponsors/xenodium][Support this work via GitHub Sponsors]] [[https://stable.melpa.org/#/chatgpt-shell][file:https://stable.melpa.org/packages/chatgpt-shell-badge.svg]] [[https://melpa.org/#/chatgpt-shell][file:https://melpa.org/packages/chatgpt-shell-badge.svg]] * ⚠️ Deprecated ⚠️ This project is now largely superseded by [[https://github.com/xenodium/agent-shell][agent-shell]], which is far more capable and actively developed. =chatgpt-shell= may continue to work, but new development happens in =agent-shell=. To create throw-away shells (as you did with =M-x chatgpt-shell=), use =M-x agent-shell-new-temp-shell=. * chatgpt-shell A multi-llm Emacs [[https://www.gnu.org/software/emacs/manual/html_node/emacs/Shell-Prompts.html][comint]] shell, by [[https://lmno.lol/alvaro][me]]. * Related packages - [[https://github.com/xenodium/ob-chatgpt-shell][ob-chatgpt-shell]]: Evaluate chatgpt-shell blocks as Emacs org babel blocks. - [[https://github.com/xenodium/ob-dall-e-shell][ob-dall-e-shell]]: Evaluate DALL-E shell blocks as Emacs org babel blocks. - [[https://github.com/xenodium/dall-e-shell][dall-e-shell]]: An Emacs shell for OpenAI's DALL-E. - [[https://github.com/xenodium/shell-maker][shell-maker]]: Create Emacs shells backed by either local or cloud services. * News chatgpt-shell goes multi model 🎉 Please sponsor the project to make development + support sustainable. | Provider | Model | Supported | Setup | |------------+---------------+-----------+----------------------------------| | Anthropic | Claude | Yes | Set =chatgpt-shell-anthropic-key= | | Deepseek | Chat/Reasoner | Yes | Set =chatgpt-shell-deepseek-key= | | Google | Gemini | Yes | Set =chatgpt-shell-google-key= | | Kagi | Summarizer | Yes | Set =chatgpt-shell-kagi-key= | | Ollama | Llama | Yes | Install [[https://ollama.com/][Ollama]] | | OpenAI | ChatGPT | Yes | Set =chatgpt-shell-openai-key= | | OpenRouter | Various | Yes | Set =chatgpt-shell-openrouter-key= | | Perplexity | Llama Sonar | Yes | Set =chatgpt-shell-perplexity-key= | Note: With the exception of [[https://ollama.com/][Ollama]], you typically have to pay the cloud services for API access. Please check with each respective LLM service. My favourite model is missing. | [[https://github.com/xenodium/chatgpt-shell/issues][File a feature request]] | [[https://github.com/sponsors/xenodium][sponsor the work]] | ** A familiar shell chatgpt-shell is a [[https://www.gnu.org/software/emacs/manual/html_node/emacs/Shell-Prompts.html][comint]] shell. Bring your favourite Emacs shell flows along. #+HTML: <img src="https://raw.githubusercontent.com/xenodium/chatgpt-shell/main/demos/cyberpunk.gif" width="80%" /> ** Swap models One shell to query all. Swap LLM provider (via =M-x chatgpt-shell-swap-model=) and continue with your familiar flow. #+HTML: <img src="https://raw.githubusercontent.com/xenodium/chatgpt-shell/main/demos/multi-model-shell.gif" width="80%" /> ** A shell hybrid =chatgpt-shell= includes a compose buffer experience. This is my favourite and most frequently used mechanism to interact with LLMs. For example, select a region and invoke =M-x chatgpt-shell-prompt-compose= (=C-c C-e= is my preferred binding), and an editable buffer automatically copies the region and enables crafting a more thorough query. When ready, submit with the familiar =C-c C-c= binding. The buffer automatically becomes read-only and enables single-character bindings. #+HTML: <img src="https://raw.githubusercontent.com/xenodium/chatgpt-shell/main/demos/compose.gif" width="80%" /> *** Navigation: n/p (or TAB/shift-TAB) Navigate through source blocks (including previous submissions in history). Source blocks are automatically selected. *** Reply: r Reply with with follow-up requests using the =r= binding. **
Excerpt of 46,187 characters
Read on GitHub
167
119
30
16
15
7
7
7
6
Kang Tu
5
4
4
4
3
3
2
2
2
2
2
Would you bet a product on this? Bounded 0–100 and slow moving.
matched fp:1ca1d27492538f40, topic:chatgpt